Get detailed Anthropic costs with model breakdown and token usage
AI agents call anthropic_costs to retrieve information from Cost Management MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool performs a read-only query of cost and usage data. No data is created, modified, deleted, or financial transactions are executed. However, severity is elevated to medium because cost/billing data is sensitive information; unauthorized access or AI agent misuse could expose financial details or enable cost estimation attacks.
From the tool's definition Tool name and description indicate it 'Get[s] detailed Anthropic costs' with 'model breakdown and token usage'—a retrieval operation with no modification or side effects.
Attacks that exploit this kind of access
Get detailed Anthropic costs with model breakdown and token usage. It is categorised as a Read tool in the Cost Management MCP MCP Server, which means it retrieves data without modifying state.
Register the Cost Management MCP server in PolicyLayer and add a rule for anthropic_costs: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Cost Management MCP. Nothing to install.
anthropic_costs is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the anthropic_costs rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for anthropic_costs. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
anthropic_costs is provided by the Cost Management MCP server (knishioka/cost-management-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →